Synthesis and reactivity of metal complexes with acyclic (amino)(ylide)carbene ligands

Article abstract of DOI:10.1002/anie.201305311

No cycle required: The straightforward synthesis of acyclic (amino)(ylide)carbene gold complexes was achieved by reaction of isocyanide gold complexes with phosphorus and arsenic ylides as well as electron-rich olefins. Their ability to form bimetallic sp

Full text of DOI:10.1002/anie.201305311

.
Angewandte
Communications
DOI: 10.1002/anie.201305311
Carbenes
Synthesis and Reactivity of Metal Complexes with Acyclic (Amino)-
(Ylide)Carbene Ligands**
Elisa Gonzꢀlez-Fernꢀndez, Jçrg Rust, and Manuel Alcarazo*
Dedicated to Professor Josꢁ M. Lassaletta
The widespread use reached by N-heterocyclic carbenes
(NHCs) as ancillary ligands for transition-metal catalysis is
a consequence of two main factors: 1) their very strong s-
electron releasing ability and 2) their relative stability, which
relies on the effective donation of electron density from both
nitrogen lone pairs to the formally empty orbital at the
carbene center.[1] However, nitrogen atoms are not unique for
this task and other entities having an electron pair located in
an orbital of appropriate symmetry might also serve the same
purpose. Thus, formal replacement of one or even both
nitrogen atoms by other main-group elements, such as O, S, or
P, has proven to be a very efficient tool to modify the donor
properties and reactivity of NHCs.[2] For this particular task,
ylides can also be seen as suitable nitrogen surrogates owing
to the non-shared electron pair on the formally negatively
charged carbon atom (Scheme 1a). Moreover, the lower
electronegativity of carbon when compared with nitrogen
should make the resulting carbenes particularly good elec-
tron-releasing ligands.
In fact, metal complexes containing such (amino)-
(ylide)carbene (AYC) ligands have been known for a long
time[3] although it was only recently that they have found
applications in catalysis.[4] Their synthesis follows two main
routes: basic treatment of the corresponding transition-metal
Scheme 1. a) Ylides as nitrogen substitutes in NHCs. b) Known entries
into metal–cyclic AYC complexes and the synthetic route developed
herein for AAYC gold complexes. AAYC: acyclic (amino)(ylide)carbene.
complexes of isocyanide units bearing a phosphonium moiety
(Scheme 1b top, route a)[3a–c] or by deprotonation of the
parent heterocyclic precursor (Scheme 1b top, route b).[5]
However, these two methods invariably deliver cyclic AYC
metal complexes while their acyclic analogues remain
unknown.[6] Considering the fundamental differences in geo-
metrical and electronic properties between cyclic and acyclic
carbenes,[7] we decided to investigate whether acyclic
(amino)(ylide)carbene (AAYC) metal complexes could also
be prepared. Outlined below is a synthetic entry that allows
the preparation of AAYC gold complexes from P-ylides, As-
ylides and interestingly, also from electron-rich olefins, such
as enamines or ene-1,1-diamines (Scheme 1b bottom). Their
structural characterization and a preliminary study of their
reactivity towards transmetalation are also described.
At first glance it seemed to us that the intermolecular
nucleophilic attack of differently substituted phosphorus
ylides on alkyl- or aryl isocyanide gold complexes could
deliver the desired AAYC gold complexes.[8] Thus, toluene
suspensions of isocyanide complexes 1 were exposed to ylides
2. In all cases consumption of 1 was observed with concom-
itant formation of two new products 3 and 4 in different ratios
depending on the electronic nature of the ylide employed and
the substituent on the isocyanide (Table 1). Multinuclear
NMR spectroscopy studies from the crude reaction mixtures
indicated that the minor component 3 only gave rise signals
attributable to the ylide fragment while 4 exhibited all the
resonance signals expected for AAYC gold complexes,
including a quite characteristic 31P NMR signal at d = 18–
22 ppm and a broad N-H peak.
[*] E. Gonzꢀlez-Fernꢀndez, J. Rust, Dr. M. Alcarazo
Max-Planck-Institut fꢁr Kohlenforschung
45470 Mꢁlheim an der Ruhr (Germany)
E-mail: alcarazo@mpi-muelheim.mpg.de
[**] Generous financial support from the Fonds der Chemischen
Industrie, the European Research Council (ERC Starting Grant to
M.A.) and SusChemSys (NRW-Ziel-2-Programm) is gratefully
acknowledged. We also thank Prof. A. Fꢁrstner for constant support
and our NMR spectroscopy and X-ray crystallography departments
for assistance.
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/anie.201305311.
11392
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2013, 52, 11392 –11395

Angewandte
Chemie
Table 1: Synthesis of AAYC gold complexes by nucleophilic attack of
phosphorus ylides on gold isocyanide complexes.[a]
3:4 ratio[b]
Yield [%]
Scheme 2. Conceivable resonance formulae of the AAYC gold com-
plexes.
Entry
R
EWG
1
2
3
4
5
6
7
8
9
Ph, 1a
COMe, 2a
COOEt, 2b
CN, 2c
CONMe2, 2d
2-py, 2e
<2:98
<2:98
<2:98
<2:98
2:98
76:24
<2:98
23:77
26:74
75:25
>98:2
85, 4a
74, 4b
88, 4c
81, 4d
98, 4e
5, 4 f
A very similar bonding situation can be found in complex
4 f (for its X-ray structure see the Supporting Information). In
contrast in 4a, conjugation between the carbonyl and the
Ph, 2 f
2,6-Me2C6H3, 1b
COMe, 2a
COOEt, 2b
CN, 2c
CONMe2, 2d
2-py, 2e
30, 4g[c]
37, 4h
25, 4i
12, 4j
83, 3e
ꢀ
carbene moieties is allowed (Figure 1). The C2 C3 distance in
this complex (1.451 ꢀ) is considerably shorter than in 4j,
ꢀ
whereas the C1 C2 bond is elongated (1.422 ꢀ), which
10
11
confirms that the most representative mesomeric form for
this compound, and other related structures, such as 4c and
4e, is III (Scheme 2).
[a] EWG=electron withdrawing group. Reaction conditions: toluene,
35–508C. [b] Determined by 31P NMR spectroscopy. [c] A conversion of
only 43% was achieved after 3 days.
In an attempt to generalize our synthetic route, we
investigated whether other ylides might also take part in the
stabilization of vicinal carbene centers. To this end, a toluene
suspension of 1a was treated with arsenic ylide 5 (Scheme 3).
Again, a smooth reaction took place giving the desired AAYC
To better ascertain the connectivities of these new
compounds, X-ray diffraction analyses of single crystals
were carried out confirming the AAYC gold complex
nature of compounds 4a–j (Figure 1 and Supporting
Information). Conversely, the isolation of side products
3a–f in pure form was not straightforward. Only crystals
of 3e could be obtained but fortunately that was enough
to unambiguously determine their constitution by NMR
spectroscopy and crystallography. This analysis showed
3e to be an ylide gold complex formed by substitution of
the isocyanide ligand originally on gold by ylide 2e
(Table 1, entry 11; Figure 1 and Supporting Informa-
tion).[9] This “parasite” ligand-exchange process leading
to the formation of 3a–f is favored by the use of
a sterically demanding isocyanide (1b) and, especially, by
using non-stabilized ylides as nucleophiles (compare
entries 6–11 of Table 1). In these cases the AAYC gold
complexes 4 f–j could still be isolated in pure form, albeit
poor yield, by iterative fractional crystallization.
ꢀ
gold complex 6 in moderate yield. The bond lengths (C2 C3
ꢀ
1.449 ꢀ, C3 O1 1.252 ꢀ) suggest a situation similar to that
Comparison of the X-ray structures of the AAYC
gold complexes depicted in Figure 1 is highly informa-
tive. Steric clash between the triphenylphosphino and
(dimethylamino)carbonyl substituents in 4j forces a com-
plete twist of the (dimethylamino)carbonyl substituents
out of the C2-C1-N1 plane (F = 90.68). Hence, this
fragment interaction with the carbene p system is largely
ꢀ
cancelled and, for this reason, the C2 C3 bond (1.513 ꢀ)
ꢀ
is longer than expected for a normal Csp2 Csp2 single bond
and better falls within the range for Csp3 Csp3 bonds.[11] As
a result, the donation of electron density from the ylide
ꢀ
ꢀ
carbon to the carbene center is maximized and the C1
C2 bond length is only 1.390 ꢀ. These structural data
indicate that compound 4j should be better described as
an intermediate between the mesomeric forms of gold
Figure 1. Crystal structures of 4a, 4e, 4j, and 3e. Solvent molecules and
hydrogen atoms, except for NH and CylideH, are removed for clarity. Thermal
carbene I and phosphoniovinyl complex II (Scheme 2). ellipsoids set at 50% probability.[10]
Angew. Chem. Int. Ed. 2013, 52, 11392 –11395
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org 11393

.
Angewandte
Communications
after work-up, a pale yellow solid (11) in 72% yield
(Scheme 4). In solution, compound 11 features the expected
1H and 13C NMR resonance signals for both a Rh-coordinated
cyclooctadiene (cod) group and a PPh3 group while the
distinctive N-H signal of 4a has vanished. Moreover,
ESI(MS) analysis showed a molecular ion signal at m/z 863
that is consistent with an structure also containing a AuCl
moiety. Single crystals were grown by slow diffusion of
pentane into a diluted solution of 11 in CH2Cl2 and X-ray
diffraction analysis revealed the formation of the expected
bimetallic species in which the Rh atom is coordinated
through the N and the lateral carbonyl group. The Au atom
remains bonded to the carbene center (see Scheme 4 for an
ORTEP plot of the molecular structure of 11).[13]
Surprisingly, this reaction is strongly influenced by the
oxidation state of the incoming metal. Treatment of
a dichloroethane solution of the organogold species 4a with
a harder Rh source, such as [{RhCp*Cl2}2] (Cp* = h-C5Me5)
and base afforded in 83% yield an orange solid that did not
seem to contain Au. Multinuclear NMR analysis again
Scheme 3. Synthesis of 6, 9, and 10, and crystal structure of 9 (solvent
molecules and hydrogen atoms, except for NH and CylideH, are
removed for clarity; thermal ellipsoids set at 50% probability).[10]
Reagents and conditions: a) 5 (1 equiv), toluene, ꢀ108C, 1 day, 40%;
b) 7 (1 equiv), toluene, 358C, 8 h, 95%; c) 8 (1 equiv), toluene, 358C,
1 h, 87%.
1
showed the disappearance of the characteristic H NMR N-
H signal from 4a; however, the 13C NMR resonance attrib-
utable to the carbene carbon atom in 4a was shifted to d =
185.1 ppm and interestingly, it appeared as a doublet of
doublets with JC,Rh = 3.3 Hz and JC,P = 36.2 Hz. These data
suggested a Au!RhIII transmetalation of the carbene and
a probable bidentate nature of the carbene ligand in the
product.[14,15] Crystals were grown by diffusion of pentane into
a saturated dichloromethane solution of this product and
described for 4a: the lateral Ph3As moiety bears a partial
positive charge while some electron density is delocalized
along the enolate side arm.
=
More interestingly, we checked if polarized C C bonds
could also participate in the stabilization of acyclic
carbenes. Thus, recognizing the “carbon ylide” char-
acter of ene-1,1-diamines and some enamines,[12] 1a
was treated under the standard conditions with
compounds 7 and 8, both containing an electron-rich
and strongly polarized double bond. The appearance
of new 13C NMR resonance signals at d = 194.6 and
189.6 ppm, which are characteristic for carbene cen-
ters coordinated to Au, supported an AAYC con-
nectivity for 9 and 10. This situation could be
confirmed by X-ray analysis of 9. Importantly, these
results broaden the structural diversity of AAYCs that
can be achieved following our synthetic procedure.
Judging from the ORTEP diagram of 9 (inset of
Scheme 3), the imidazolium–enolate dipolar form 9’ is
ꢀ
ꢀ
dominant: the C2 C3 (1.435 ꢀ) and the C3 O1 bond
lengths (1.266 ꢀ) are respectively the shortest and
longest ones measured along the complete series of
organogold complexes, whereas the C1-C2-C4-N3
torsion angle of 75.98 prevents any efficient overlap
between the imidazolium cation and the carbene
p system (Scheme 3).
Finally, we envisioned the possibility to prepare
heterobinuclear complexes making use of the addi-
tional functional groups present at the acyclic carbene
side arms. In fact, the formation of metal chelates
should be favored after deprotonation of the NH unit
adjacent to the carbene. Hence, treatment of a THF
solution of 4a with 1 equivalent of KOMe and
Scheme 4. Synthesis of 11–13 and crystal structures of 11 and 12 (solvent
molecules and hydrogen atoms are removed for clarity; thermal ellipsoids set at
50% probability).[10] Reagents and conditions: a) 4a (1 equiv), [{Rh(cod)Cl}2]
(0.5 equiv), KOMe (1 equiv), THF, 58C, 12 h; 72%; b) 4a (1 equiv),
[{RhCp*Cl2}2] (0.5 equiv), Et3N (15 equiv), dichloroethane, 508C, 4 days; 83%;
c) 4a (1 equiv), [{RhCp*Cl2}2] (0.5 equiv), KOMe (0.5 equiv), THF, 58C, 12 h;
41%; d) 4a (1 equiv), [{Ru(cym)Cl2}2] (0.5 equiv), Et3N (15 equiv), dichloro-
0.5 equivalents of [{RhCl(cod)}2] at 58C afforded, ethane, 508C, 1 day; 59%. cym: p-cymene.
11394 www.angewandte.org
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2013, 52, 11392 –11395

Angewandte
Chemie
[5] a) S. Nakafuji, J. Kobayashi, T. Kawashima, Angew. Chem. 2008,
subsequently its structure was determined by X-ray diffrac-
tion as 12 (Scheme 4). As predicted from the spectroscopic
analysis, the AuCl moiety is not present in 12, thus confirming
the ability of 4a to transfer the organic ligand. Moreover, the
nitrogen atom has been deprotonated and the resulting
monoanionic ligand now coordinates the [RhCp*Cl]+
moiety in a bidentate fashion through the central carbon
atom and the lateral carbonyl group.
Interestingly, the same ligand-transfer reactivity was
observed upon exposure of 4a to [{Ru(cym)Cl2}2] affording
compound 13. Considering that the synthesis of acyclic
carbene metal complexes through the isocyanide route is
basically limited to AuI, PdII, and PtII isocyanide precursors,[16]
this reactivity seems to offer access to complexes of different
metals provided that chelation of the incoming metal is
possible.
120, 1157; Angew. Chem. Int. Ed. 2008, 47, 1141; b) M. Asay, B.
Donnadieu, A. Baceiredo, M. Soleilhavoup, G. Bertrand, Inorg.
Chem. 2008, 47, 3949; c) A. Fꢂrstner, M. Alcarazo, K. Radkow-
ski, C. W. Lehmann, Angew. Chem. 2008, 120, 8426; Angew.
Chem. Int. Ed. 2008, 47, 8302; d) J. Kobayashi, S. Nakafuji, A.
Yatabe, T. Kawashima, Chem. Commun. 2008, 6233.
[6] Carbenes stabilized by an ylide moiety and an oxygen substitu-
ent have been prepared by reaction of [M(CO)6] (M = Cr, Mo,
W) with P ylides and further alkylation/silylation of the resulting
intermediate; see: a) W. C. Kaska, D. K. Mitchell, R. F. Reich-
elderfer, W. D. Korte, J. Am. Chem. Soc. 1974, 96, 2847; b) S.
Voran, H. Blau, W. Malish, U. Schubert, J. Organomet. Chem.
1982, 232, c33 – c40; c) N. E. Kolobova, L. L. Ivanov, O. S.
Zhvanko, I. N. Chechulina, A. S. Batsanov, Y. T. Struchkov, J.
Organomet. Chem. 1982, 238, 223.
[7] a) J. Vignolle, X. Cattoꢄn, D. Bourissou, Chem. Rev. 2009, 109,
3333; b) L. M. Slaughter, ACS Catal. 2012, 2, 1802.
[8] For the synthesis of acyclic diamino carbenes and NHCs by
nucleophilic attack to gold isocyanide complexes see for
example: a) C. Bartolomꢅ, Z. Ramiro, D. Garcꢆa-Cuadrado, P.
Pꢅrez-Galꢇn, M. Raducan, C. Bour, A. M. Echavarren, P.
Espinet, Organometallics 2010, 29, 951; b) A. S. K. Hashmi, C.
Lothschꢂtz, C. Bçhling, F. Rominger, Organometallics 2011, 30,
2411; c) A. S. K. Hashmi, C. Lothschꢂtz, C. Bçhling, T. Hengst,
C. Hubbert, F. Rominger, Adv. Synth. Catal. 2010, 352, 3001;
d) A. S. K. Hashmi, T. Hengst, C. Lothschꢂtz, F. Rominger, Adv.
Synth. Catal. 2010, 352, 1315; e) A. S. K. Hashmi, Y. Yu, F.
Rominger, Organometallics 2012, 31, 895; f) S. Handa, L. M.
Slaughter, Angew. Chem. 2012, 124, 2966; Angew. Chem. Int. Ed.
2012, 51, 2912; g) R. Manzano, F. Rominger, A. S. K. Hashmi,
Organometallics 2013, 32, 2199; h) M. C. Blanco-Jaimes, C. R. N.
Bçhling, J. M. Serrano-Becerra, A. S. K. Hashmi, Angew. Chem.
2013, 125, 8121; Angew. Chem. Int. Ed. 2013, 52, 7963.
[9] For a previous synthesis of compound 3e see: J. Vicente, M. T.
Chicote, M. C. Lagunas, Inorg. Chem. 1993, 32, 3748.
[10] CCDC 937966 (3e), 937968 (4a), 937970 (4c), 937969 (4e),
937972 (4 f), 937967 (4j), 937971 (6), 937974 (9), 945351 (11),
and 937973 (13) contain the supplementary crystallographic data
for this paper. These data can be obtained free of charge from
The Cambridge Crystallographic Data Centre via www.ccdc.
cam.ac.uk/data_request/cif.
[11] F. H. Allen, O. Kennard, D. G. Watson, L. Brammer, A. G.
Orpen, R. Taylor, J. Chem. Soc. Perkin Trans. 2 1987, S1 – S19.
[12] For general Reviews on ene-1,1-diamines see: a) W. Kantlehner,
Science of Synthesis, Vol. 24 (Ed.: A. de Meijere), Thieme,
Stuttgart 2006, p. 571; b) P. A. Keller, J. Morgan Science of
Synthesis, Vol. 24 (Ed.: A. de Meijere), Thieme, Stuttgart 2006,
p. 707.
In summary, the synthesis of unprecedented AAYC Au
complexes has been achieved by reaction of gold isocyanides
with phosphorus or arsenic ylides. In addition, the reaction
could be applied with excellent yields to “carbon ylides”, such
as ene-1,1-diamines and enamines. Finally, the ability of the
newly prepared AAYC Au complexes to form bimetallic
species and to act as ligand-transfer reagents was established
by reaction with appropriate Rh or Ru sources. Whether the
mono- and bimetallic complexes described herein could find
applications in catalysis is under investigation.
Received: June 20, 2013
Published online: September 3, 2013
Keywords: acyclic carbenes · gold · ligand design ·
.
transmetalation · ylides
[1] a) J. A. Arduengo, Acc. Chem. Res. 1999, 32, 913; b) D.
Bourissou, O. Guerret, F. P. Gabaꢁ, G. Bertrand, Chem. Rev.
2000, 100, 39; c) W. A. Herrmann, Angew. Chem. 2002, 114,
1342; Angew. Chem. Int. Ed. 2002, 41, 1290; d) F. E. Hahn,
Angew. Chem. 2006, 118, 1374; Angew. Chem. Int. Ed. 2006, 45,
1348; e) M. Melaimi, M. Soleilhavoup, G. Bertrand, Angew.
Chem. 2010, 122, 8992; Angew. Chem. Int. Ed. 2010, 49, 8810.
[2] a) J. A. Arduengo, J. R. Goerlich, W. J. Marshall, Liebigs Ann./
Recl. 1997, 365; b) R. W. Alder, C. P. Butts, A. G. Orpen, J. Am.
Chem. Soc. 1998, 120, 11526; c) D. Martin, A. Baceiredo, H.
Gornitzka, W. W. Schoeller, G. Bertrand, Angew. Chem. 2005,
117, 1728; Angew. Chem. Int. Ed. 2005, 44, 1700; d) A. S. K.
Hashmi, C. Lothschꢂtz, K. Graf, T. Hꢃffner, A. Schuster, F.
Rominger, Adv. Synth. Catal. 2011, 353, 1407.
[13] For a one-pot tandem Au/Rh-catalyzed process see: M. M.
Hansmann, A. S. K. Hashmi, M. Lautens, Org. Lett. 2013, 15,
3226.
[14] Y. Shi, S. A. Blum, Organometallics 2011, 30, 1776.
[15] An NHC ligand transfer from Au to Pd has already been
described: S. Liu, C. Lee, C. Fu, C. Chen, Y. Liu, C. J. Elsevier, S.
Peng, J. Chen, Organometallics 2009, 28, 6957.
[3] a) G. Facchin, R. Campostrini, R. A. Michelin, J. Organomet.
Chem. 1985, 294, c21 – c25; b) R. A. Michelin, G. Facchin, D.
Braga, P. Sabatino, Organometallics 1986, 5, 2265; c) R. A.
Michelin, M. Mozzon, G. Facchin, D. Braga, P. Sabatino, J.
Chem. Soc. Dalton Trans. 1988, 1803; d) M. Tamm, F. E. Hahn,
Coord. Chem. Rev. 1999, 182, 175.
[16] A. S. K. Hashmi, C. Lothschꢂtz, C. Bçhling, F. Rominger,
Organometallics 2011, 30, 2411. See also reference [8].
[4] M. Alcarazo, T. Stork, A. Anoop, W. Thiel, A. Fꢂrstner, Angew.
Chem. 2010, 122, 2596; Angew. Chem. Int. Ed. 2010, 49, 2542.
Angew. Chem. Int. Ed. 2013, 52, 11392 –11395
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org 11395